Intel Arc Pro B370 vs NVIDIA Jetson Orin Nano Super Comparison

# Intel Arc Pro B370 vs NVIDIA Jetson Orin Nano Super

The Intel Arc Pro B370 and NVIDIA Jetson Orin Nano Super are both compact, low-power graphics solutions aimed at different segments of the embedded and mobile market. The Intel part is an integrated graphics processor built on Panther Lake silicon, while the NVIDIA part is a standalone module featuring the GA10B chip. Both carry a 25 W TDP and support DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4, which makes them superficially similar in API compatibility. However, the underlying architecture, memory configuration, and compute capabilities diverge sharply. The recorded data shows that the Intel Arc Pro B370 offers substantially higher raw graphics throughput, while the NVIDIA Jetson Orin Nano Super brings dedicated tensor cores and a fixed 8 GB memory pool to the table. This analysis draws only from the database entries for these two parts.

Head-to-Head Benchmarks

The database contains no direct benchmark scores for either GPU, so the comparison relies on the recorded compute and rendering specifications. The Intel Arc Pro B370 delivers 6.144 TFLOPS of FP32 performance, which is roughly three times the 2.089 TFLOPS of the Jetson Orin Nano Super. In FP16 workloads, the Intel part reaches 12.29 TFLOPS with a 2:1 ratio, while the NVIDIA module produces 4.178 TFLOPS under the same ratio. That means the Intel GPU holds a 2.94x advantage in both FP32 and FP16 peak throughput.

Pixel throughput follows a similar pattern. The Intel Arc Pro B370 sustains 48.00 GPixel/s, compared to 16.32 GPixel/s for the Jetson Orin Nano Super, a 2.94x margin. Texture fill rates are even more lopsided: 96.00 GTexel/s for Intel versus 32.64 GTexel/s for NVIDIA, again a 2.94x difference. These ratios are consistent across all four metrics because the Intel part uses 1280 shading units, 40 texture mapping units, and 20 ROPs, while the NVIDIA part uses 1024 shading units, 32 TMUs, and 16 ROPs. The Intel configuration has exactly 25% more units in each category, yet its clock advantage amplifies that into a near-triple performance lead.

The memory subsystem tells a different story. The Jetson Orin Nano Super has 8 GB of LPDDR5 memory on a 128-bit bus, delivering 102.4 GB/s of bandwidth. The Intel Arc Pro B370 uses system shared memory with a bus width and bandwidth listed as system dependent, so no fixed bandwidth figure exists in the database. For scenarios where dedicated memory bandwidth matters, such as large dataset processing or sustained tensor workloads, the NVIDIA module has a defined advantage on paper. The Intel part's memory performance cannot be quantified from the recorded data, as it depends entirely on the host platform.

Architecture Differences

The Intel Arc Pro B370 is built on the Xe3-LPG architecture, part of the Arc Graphics-WM generation for Panther Lake. It uses a 3 nm process node manufactured by Intel. The NVIDIA Jetson Orin Nano Super uses the Ampere architecture, specifically the GA10B chip from the Tegra generation, on an 8 nm Samsung process. The die size for the NVIDIA part is 200 mm², while the Intel die size is listed as unknown. Transistor counts are unknown for both parts.

The Intel GPU includes 10 ray tracing cores, which the NVIDIA module lacks entirely. Conversely, the NVIDIA module includes 32 tensor cores, a feature absent from the Intel Arc Pro B370. This is a fundamental architectural split: Intel targets real-time ray tracing and rasterization, while NVIDIA focuses on AI acceleration through its tensor core array. The Intel part also has a higher shading unit count at 1280 versus 1024, along with more TMUs and ROPs.

Clock behavior differs as well. The Intel Arc Pro B370 has a base clock of 300 MHz and a boost clock of 2400 MHz. The NVIDIA Jetson Orin Nano Super has no base or boost clock listed in the database; instead, its memory clock is specified as 800 MHz with 6.4 Gbps effective data rate. The Intel GPU's memory clock is system shared, meaning it inherits whatever the host platform provides.

Both parts are integrated-class solutions with an IGP slot width. The Intel part uses an IGP bus interface with no power connectors, while the NVIDIA module uses a PCIe 4.0 x4 interface. The NVIDIA module has physical dimensions of 70 mm by 45 mm, roughly 2.8 inches by 1.8 inches, while the Intel part lists no dimensions since it is integrated into the host processor. Display outputs for both are listed as portable device dependent, and both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Where Each One Wins

The Intel Arc Pro B370 wins in every raw graphics throughput category recorded in the database. Its FP32 and FP16 compute are nearly three times higher, its pixel fill rate is nearly three times higher, and its texture fill rate is nearly three times higher. The presence of 10 ray tracing cores gives it a hardware capability that the NVIDIA module simply does not offer. For any workload that depends on rasterization, shader compute, or ray tracing, the Intel part is the clear choice based on the recorded specifications.

The NVIDIA Jetson Orin Nano Super wins in the AI and machine learning domain. Its 32 tensor cores provide dedicated hardware for neural network inference and training tasks, something the Intel part cannot match because it has no tensor core equivalents. The NVIDIA module also has a fixed 8 GB LPDDR5 memory pool with 102.4 GB/s of bandwidth, which is a concrete, platform-independent resource. The Intel part relies on system shared memory, so its available capacity and bandwidth vary with the host platform. For developers who need predictable memory behavior and tensor acceleration, the NVIDIA module offers capabilities that the Intel part lacks.

The NVIDIA module also has a defined physical footprint. Its 70 mm by 45 mm dimensions make it a small, self-contained compute board. The Intel part, being an IGP, has no standalone form factor and must be paired with a Panther Lake host processor. The NVIDIA module's PCIe 4.0 x4 interface allows it to be installed in a variety of carrier boards, while the Intel IGP is fixed to its host platform.

The Verdict

The data indicates a clear split by workload type. For traditional graphics rendering, compute shaders, and ray tracing, the Intel Arc Pro B370 is substantially stronger. Its 6.144 TFLOPS FP32 and 12.29 TFLOPS FP16 performance, combined with 10 ray tracing cores, make it the higher-performing GPU in every measured graphics category. The near-triple advantage in pixel and texture fill rates means frame rendering and texture-heavy scenes will benefit significantly from the Intel solution.

For AI inference, embedded edge computing, and tensor-based workloads, the NVIDIA Jetson Orin Nano Super is the appropriate choice. Its 32 tensor cores are a dedicated resource that the Intel part cannot emulate. The fixed 8 GB LPDDR5 memory with 102.4 GB/s bandwidth provides a consistent memory environment, which is critical for model deployment and real-time inference. The 200 mm² die size and 8 nm Samsung process are less advanced than Intel's 3 nm node, but the architectural focus on AI acceleration compensates for the process disadvantage in neural network tasks.

The NVIDIA module also carries a launch MSRP of 249 USD, while the Intel part has no listed launch MSRP since it is integrated into a host processor. Both parts sit at the 50th percentile among all GPUs in the database, though neither has any recorded benchmark scores or nearest rivals. Buyers should choose based on whether their priority is graphics throughput or tensor acceleration. The Intel Arc Pro B370 delivers more raw compute and rendering capability, but the NVIDIA Jetson Orin Nano Super provides specialized AI hardware and a self-contained form factor.

FAQ

Q: Which GPU has higher FP32 performance?

A: The Intel Arc Pro B370 delivers 6.144 TFLOPS of FP32 performance, which is 2.94x higher than the 2.089 TFLOPS of the NVIDIA Jetson Orin Nano Super.

Q: Does either GPU support ray tracing?

A: Yes, the Intel Arc Pro B370 includes 10 ray tracing cores. The NVIDIA Jetson Orin Nano Super has no ray tracing cores listed in the database.

Q: Which GPU has tensor cores?

A: The NVIDIA Jetson Orin Nano Super includes 32 tensor cores. The Intel Arc Pro B370 has no tensor cores listed.

Q: How much memory does each GPU have?

A: The NVIDIA Jetson Orin Nano Super has 8 GB of LPDDR5 memory on a 128-bit bus with 102.4 GB/s bandwidth. The Intel Arc Pro B370 uses system shared memory with size, type, and bus width all listed as system shared, and bandwidth listed as system dependent.

Q: What are the process nodes for these GPUs?

A: The Intel Arc Pro B370 uses a 3 nm process from Intel. The NVIDIA Jetson Orin Nano Super uses an 8 nm process from Samsung.

Q: What is the TDP of each GPU?

A: Both the Intel Arc Pro B370 and the NVIDIA Jetson Orin Nano Super have a TDP of 25 W.

Specification Differences

| Specification | Intel Arc Pro B370 | NVIDIA Jetson Orin Nano Super |

|---|---|---|

| Architecture | Xe3-LPG | Ampere |

| Process Node | 3 nm | 8 nm |

| Foundry | Intel | Samsung |

| Die Size | Unknown | 200 mm² |

| Base Clock | 300 MHz | Not listed |

| Boost Clock | 2400 MHz | Not listed |

| Memory Size | System Shared | 8 GB |

| Memory Type | System Shared | LPDDR5 |

| Memory Bus Width | System Shared | 128 bit |

| Memory Bandwidth | System Dependent | 102.4 GB/s |

| Shading Units | 1280 | 1024 |

| TMUs | 40 | 32 |

| ROPs | 20 | 16 |

| Ray Tracing Cores | 10 | Not listed |

| Tensor Cores | Not listed | 32 |

| Pixel Rate | 48.00 GPixel/s | 16.32 GPixel/s |

| Texture Rate | 96.00 GTexel/s | 32.64 GTexel/s |

| FP32 Performance | 6.144 TFLOPS | 2.089 TFLOPS |

| FP16 Performance | 12.29 TFLOPS (2:1) | 4.178 TFLOPS (2:1) |

| Bus Interface | IGP | PCIe 4.0 x4 |

| Dimensions | Not listed | 70 mm by 45 mm |

| Launch MSRP | Not listed | 249 USD |

| Release Date | 2026-01-26 | 2024-12-16 |

| Predecessor | HD Graphics-WM | Not listed |